No other period in human life matches the magnitude and astonishing increase in neural functional[unreadable] capacity as the transition from neonate to infant to toddler. Within a month or two following birth, the infant[unreadable] begins to smile and around first birthday begins to speak. Soon thereafter he coordinates speech and[unreadable] gesture in order to garner the attention of others. By the time he is two, he can speak in short sentences and[unreadable] has a rich repertoire of socio-emotional responding.[unreadable] Of all autism biology, the greatest mystery is what neural malfunction fails to provide this leap in[unreadable] social, emotional and language capacity.[unreadable] The only safe and powerful method for observing human neural function directly is functional MRI.[unreadable] While there are now thousands of fMRI studies in mature humans, not a single fMRI experiment has[unreadable] attempted to discover the neurofunctional bases of this incredible achievement in typical infants or its[unreadable] profound failure in the autistic infant.[unreadable] We have performed ground-breaking pilot fMRI studies of typical and autistic toddlers mapping for the[unreadable] first time the regions of the brain that are involved in responding to language, social and emotion sounds.[unreadable] We have done this in the only way feasible: while the toddler is in natural sleep and not moving. This method[unreadable] of mapping neural function and malfunction allows testing to be done comparably regardless of[unreadable] developmental age or level of mental or behavioral ability; infants at all ages and levels of functioning can be[unreadable] studied equivalently.[unreadable] We therefore propose the first-of-its-kind fMRI study of infants at-risk for autism at 12 and 24 months.[unreadable] Our proposed natural sleep fMRI experiments will be the first to map the neural systems activated by social,[unreadable] emotional and language information in the autistic, developmental^ delayed, and typically developing infant.[unreadable] Abnormal fMRI activation patterns will be identified that characterize and predict social emotional and[unreadable] language behavioral deficits as measured in clinical assessment and diagnosis Core B. Behavioral deficits[unreadable] in these three areas are among the first red flags signaling possible autism in the infant, and through our[unreadable] studies we will identify the first neurofunctional biomarkers of early development in autism. Identification of[unreadable] these neural functional biomarkers of early development in autism will enable us through collaboration with[unreadable] the MRI Project 1, genetic Projects 3 and 4, the Treatment Core C and the Integrated Biostatistics Core D, to[unreadable] perform the first ever analyses that will establish relationships between brain function on the one hand and[unreadable] neuroanatomy, gene expression and genetic polymorphisms, clinical variables and treatment effectiveness[unreadable] on the other.[unreadable]